PSY 370 Lecture Notes - Lecture 37: British Rail Class 37, Psy, Factor Analysis
Document Summary
You might have a reliable test that is just not relating well to your outcomes. ***reliability is a necessary but not sufficient condition for validity. You cannot have a criterion-valid test that is not reliable. You can have a reliable test that is not criterion-valid. Something about the way we"ve measured the criterion might be the problem. *validity is about our interpretations of test scores. If the test really measures what we say it measures, these should go together but that"s not always the case. We may write a well-designated, content valid test that doesn"t predict well. We might also write a test that predicts well but doesn"t appear to be related to our construct. Culture, translation, etc. raise the possibility that a test might predict well for one group but not another. Erroneously predicting a person"s score has serious consequences when the test is being used for decision making.
